New pages indexing without the www.

[LukeC]

We are now having new pages indexing like this:

fluffywidgets.com/best-widgets-ever.html

instead of

www.fluffywidgets.com/best-widgets-ever.html

The only thing we can think it might be is we were doing 301 redirects to change the file names, but why would it leave out the www?

301 redirects appear to be causing havoc with google for us, we are getting more pages going in on a daily basis but now we have noticed our index page disappear.

[jimh009]

I've seen this happen on some of my static .html pages that use no redirects. It happens without any reason and is very random - only a page here and there is affected. Moreover, it doesn't seem to affect rankings at all. Expect it to last about a month or so.

[vincevincevince]

You can use mod_rewrite to ensure that without a www. you serve a 302 to with a www - it works fine for me.

[Marcia]

I'm seeing that a lot now, even when there are no links going to non-wwww, and even on new domains. At first I used a 301 on my sites where I caught it first, figuring that it was a slip-up, but it's too widespread for that.

[steveb]

If you use relative links, it only takes one link to a non-www page to send the bot through the whole non-www loop.

LukeC, ditch the relative links and make your domain linking consistent.

[Marcia]

Steve, linking should always be done consistently. But in numerous cases that have been observed, it has nothing at all to do with how the webmaster is doing links.

The important thing is, that these are showing up where there are ONLY absolute links done one way, with the www, and ONLY links that the webmaster knows about, which are also absolute links, with the www. Even brand new domains that no one whatsoever is linking to, except maybe the web designer - with one link, absolute with the www.

They are URL only, which means that the content isn't indexed with it, and are without a shadow of a doubt coming from someplace else - from a link someplace else that is beyond the webmaster's knowledge or control.

I believe that Google is getting the URLs, wirhout the www, from a database of whois records that has access to the info as soon as a registered site is put on hosting.

I'm talking about the root domain.